A brain consists of a plurality of neurons that interact by exchanging chemical messengers. Each neuron generates neurochemicals, referred to as neurotransmitters; neurotransmitters act at sites on the cellular membrane of a neuron, the sites being referred to as receptors. Receptors are associated with either ion channels through the cellular membrane or secondary neurochemical messenger systems. By contrast, reuptake sites are molecular complexes which transport chemicals across the cellular membrane of a neuron. When a neurotransmitter has served its function, it is removed from the vicinity of the receptor by being bound to a reuptake site which transports the neurotransmitter to the interior of the neuron.
Just as there are many specialized neurons in the brain, there are also a variety of neurotransmitters, associated receptors, and reuptake sites. The distribution of specialized neurons depends upon the particular organism under study, and the state of health of that organism.
A neuron can be classified according to the type of neurotransmitter that it uses to communicate with other neurons. Certain types of neurons can be found predominantly in particular regions of the brain. For example, the striatal region of a mammalian brain is innervated by neurons using dopamine as a neurotransmitter. The striatum also contains a large number of non-dopaminergic neurons that have dopamine receptors. Certain compounds, such as cocaine, have a preferential affinity for dopamine reuptake sites, and therefore tend to bind to such reuptake sites. The effect of a molecule such as cocaine upon a dopamine reuptake site is to inhibit reuptake of the neurotransmitter dopamine, leaving more dopamine available in the vicinity of the dopamine receptors.
In certain neurological diseases, such as Parkinson's disease, distinct groups of neurons lose their normal physiological functioning. Consequently, the abnormal neurons may behave differently in the presence of some neurotransmitters, and may also produce neurotransmitters in a manner that differs from a healthy neuron.
The major neurotransmitters, dopamine, norepinephrine, and serotonin, are referred to collectively as the monoamine neurotransmitters. Many neurons have receptors adapted to receive at least one of these neurotransmitters. Parkinson's disease is caused by the degeneration of some of the dopaminergic neurons in the brain. The neurons lost in Parkinson's disease have a large number of dopamine reuptake sites; cocaine and chemical analogs of cocaine have an affinity for such reuptake sites.
A radioisotope is commonly incorporated in molecules that have a demonstrated binding affinity for a particular type of neuroreceptor, and such molecules are commonly used as neuroprobes. The localization of neuroprobes can be used to find specialized neurons within particular regions of the brain. It is also known that a neurological disease can be detected by observing abnormal binding distributions of a neuroprobe. Such abnormal binding distributions can be observed by incorporating a radionuclide within each molecule of the neuroprobe with a high binding affinity for the particular reuptake sites of interest. Then, an imaging technique can be used to obtain a representation of the in vivo spatial distribution of the reuptake sites of interest.
In single photon emission computed tomography (SPECT) imaging, the most commonly used radionuclides are heavy metals, such as .sup.99m Tc. Heavy metals are very difficult to incorporate into the molecular structure of neuroprobes because such probes are relatively small molecules (molecular weight less than 400).
In positron emission tomography (PET), the radiohalide .sup.18 F (fluorine) is commonly used as a substitute for H (hydrogen) in radiopharmaceuticals because it is similar in size. Not all halogens will work, however. For example, I (iodine) is much larger than both H and F, being approximately half the size of a benzene ring. However, due to the small size of typical radiopharmaceuticals for use as neuroprobes, the presence of iodine markedly changes the size of the compound, thereby altering or destroying its biological activity.
In addition, the presence of iodine in a neuroprobe tends to increase its lipophilicity, and therefore increases the tendency of the neuroprobe to engage in non-specific binding. For example, paroxetine is a drug with high affinity and selectivity for serotonin reuptake sites, and [3H]paroxetine has been shown in rodents to be a useful in vivo label (Scheffel, U. and Hartig, PR. J. Neurochem., 52: 1605-1612, 1989). However, several iodinated analogs of this compound with iodine attached at several different positions had unacceptably low affinity, in fact being one tenth of the affinity of the parent compound. Furthermore, when the iodinated compound was used as an in vivo radiolabled neuroprobe, non-specific binding activity was found to be so high that no measurable portion of the brain uptake appeared to be specifically bound to the serotonin reuptake site. Thus, the iodinated form of paroxetine is not useful as an in vivo probe.
The addition of iodine to a neuroprobe can unfavorably alter its biological properties. For example, tomoxetine has high affinity and selectivity for norepinephrine reuptake sites. However, when tomoxetine is iodinated, e.g. to form R-4-iodotomoxetine, the resulting labeled compound has low affinity for such reuptake sites, and relatively high affinity for serotonin reuptake sites. In vivo labeling studies have shown that it is an unacceptably poor probe even for the serotonin reuptake sites because it exhibits low total brain uptake and immeasurably low specific uptake.
An iodinated compound can be useful as an in vitro probe, but may be useless as an in vivo probe, because an in vivo probe must meet the requirements associated with intravenous administration of the probe to a living subject. Reasons for the loss of in vivo utility include the fact that the compound may be metabolized too quickly, that it may not cross the blood-brain-barrier, and that it may have high non-specific uptake into the lipid stores of the brain. In vitro homogenate binding studies remove these obstacles by isolating the brain tissue from hepatic metabolic enzymes, by homogenizing the brain tissue so as to destroy the blood-brain-barrier, and by diluting the brain tissue so as to decrease the concentration of lipids in the assay tube. Accordingly, it cannot be assumed that a probe will be useful in both in vivo and in vitro modalities.
An in vivo SPECT probe was developed by iodinating cocaine. However, this probe shows a binding affinity and specificity no better than cocaine itself, which is inadequate for purposes of SPECT imaging.